Dear participants, we hope your predictions are going well (and yourself of course).
Given your suggestion during the webinar to establish a leaderboard, we came up wit the following proposal:
In order to enhance competition,
A leaderboard using the final test set, but showing only ranks (to avoid overfitting given the small test sets), will be put up .
Participants will be able to submit models once a week for each subchallenge and ranks will be updated at the end of the week, for 3 weeks.
So 3 total submissions and 3 updated ranks.
Scoring scheme will not be revealed
What do you think?
Thanks
The DREAM Malaria challenge organizers